This invention relates to a semi-quantitative method for determining small quantities of water (generally less than about 1%v) in oil. The invention particularly relates to the determination of water in hydraulic and lubricating oils.
In the past, great emphasis has been placed on ensuring that the condition of new hydraulic and automotive oils is satisfactory, i.e., that their composition and viscosity are correct, and that their dirt and water contents are negligible. However, once in service, these oils are often subjected to contamination from both dirt and water as well as subject to slow deterioration due to additive depletion and oxidation. Although these developments can be readily assessed in the laboratory, it is time consuming and expensive to convey samples to and from a service laboratory. It is therefore desirable to devise equipment suitable for simple, cheap, rapid and accurate on-the-spot monitoring of the condition of mineral-oil-based oils from which their expected future service life might be assessed.
For years, many automotive and industrial equipment users have used a so-called "crackle" test (often referred to as the snap, pop and crackle test) to check for water in lubricants. The test involves dropping a sample of the questionable oil on a hot surface. Depending on the temperature involved, a dry oil will smoke, while a wet oil will "crackle". A safer method than using a hot surface to drop the oil on, is one where one or two drops of oil are placed on aluminum foil and a match flame is slowly applied to the under surface of the foil. Other methods include the use of a spoon, aluminum weighing dish, etc. The above methods, however, are non-quantitative, and only indicate the presence of water.
A study of several commercially available test kits has revealed that some were totally inadequate for use in determining the water content of oils, while others were too time consuming.
One such kit was found to be non-quantitative and very crude. It comprises heating a "hot plate" (bent 18 gauge steel clipped to and held above a small spirit lamp heater) to above 100.degree. C. and placing a drop of oil in an indent in this hotplate. The amount of moisture present is indicated by the intensity of crackling and frothing as the water boils off. However, since degeneration of hydraulic equipment can begin at quite low water levels, it is essential that these levels be accurately determined.
The "Oil Type Speedy Moisture Tester", manufactured by Thomas Ashworth and Co. Ltd., Burnley, Lancs, U.K. was found to be lacking in accuracy and general usefulness. In common with most of the kits examined the "Speedy" relies upon a water finding reagent, calcium carbide in this case, reacting with, and producing a pressure rise in direct proportion to, water content. The water content of the oil can be read directly from a pre-calibrated gauge. For the detection of water in oil it is recommended that the sample be placed in the apparatus in the presence of a dry sand adsorbent (presumably this is to make the water more accessible for reaction with the non-soluble carbide). When the procedure outlined in the "Speedy" literature was followed, inconsistent and inaccurate results were obtained.
We have developed a method which overcomes some of the limitations of prior art water test kits.